Context: Several cohort studies have reported U-shaped or inverse J-shaped associations between circulating 25-hydroxyvitamin D [25OHD] and clinical outcomes.
Objective: We aimed to investigate in cardiac surgical patients the association of preoperative 25OHD and 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] levels with the risk of major adverse cardiac and cerebrovascular events (MACCE).
Design: A prospective cohort study of adult cardiac surgical patients in 2012-2013 was used.
Setting: The study was conducted at the Heart and Diabetes Center North Rhine-Westphalia, Germany.
Patients: A total of 3371 adult patients participated in the study.
Intervention: None Measurements: The main outcome measure was MACCE until discharge. We categorized vitamin D metabolite levels into subgroups and performed multivariable-adjusted logistic regression analysis to estimate odds ratios (ORs) of MACCE. Moreover, we performed multiple regression analysis to assess the association of 25OHD and circulating 1,25(OH)2D3 with preoperative parameters.
Results: As compared with patients in the 25OHD reference category (75-100 nmol/L), the multivariable-adjusted odds ratios (OR) of MACCE was significantly higher in patients with deficient 25OHD levels (< 30 nmol/L) (OR = 2.06 [95%CI: 1.24-3.43]), but was comparable in patients with 25OHD levels > 100 nmol/L (OR = 1.16 [95% CI: 0.56-2.37]). Poor kidney function was an important predictor of high 25OHD (>100 nmol/L) and low 1,25(OH)2D3 levels. 1,25(OH)2D3 was not independently associated with the incidence of MACCE.
Conclusions: In cardiac surgical patients, deficient but not high 25OHD levels are independently associated with the risk of MACCE. Cohort studies should consider potential interrelationships between kidney function, circulating vitamin D metabolite levels, and clinical outcome.